Breastfeeding is an important health-promoting behavior. Human milk is the ideal diet for all infants, optimizes intellect, and provides protection against infectious and atopic diseases in childhood as well as decreasing risks for obesity, hypertension and other chronic diseases. Infants with the highest risk of life-long disability, very low birthweight (VLBW) preterm infants, are breastfed at some of the lowest rates in the US. Maternal milk is not always available, and pasteurized donor human milk is an alternative that requires investigation. Whether donor milk conveys health and developmental advantages similar to those bestowed by maternal milk is unknown. By determining the effects of donor milk on health and developmental outcomes when compared to preterm infant formula, we seek to optimize outcomes in this fragile population. Term and late-preterm infants admitted to NICUs are breastfed at lower rates than healthy term infants. Understanding factors predictive of breastfeeding success in these populations is essential to designing interventions. This application describes a 5-year training plan to develop skills necessary for success as an independent neonatal epidemiology investigator. The PI is guaranteed at least 80% protected time to devote to research during this award. Extensive support has been given to the candidate by senior investigators both internal and external to the proposed research, and will continue. Drs. Saftlas and Morriss will sponsor and co-sponsor the Pi's training, and their respective expertise in perinatal epidemiology and large-database analysis will guide the Pi's efforts. The hypothesis of our donor milk research is that donor human milk offers health and developmental advantages similar to maternal milk in VLBW infants. The Pi's preliminary retrospective data show similar developmental outcomes in donor milk and maternal milk fed infants, leading to the following aims: 1) To conduct a randomized trial of donor human milk vs. preterm infant formula, to determine if short and medium-term health and developmental outcomes are enhanced by donor milk compared to formula; 2) To retrospectively compare growth through 18-22 months in preterm infants fed donor human milk compared with those fed maternal milk, with the hypothesis that growth is equivalent; 3) To determine factors predictive of breastfeeding success in the term and late-preterm populations admitted to NICUs, in order to design interventions to eradicate barriers.